Multi-Year Synthesis with a Focus on Pcbs and Hg
Total Page:16
File Type:pdf, Size:1020Kb
CLEAN WATER / RMP NUMBER 773 DECEMBER 2015 Sources, Pathways and Loadings: Multi-Year Synthesis with a Focus on PCBs and Hg Prepared by Lester J. McKee, Alicia N. Gilbreath, Jennifer A. Hunt, Jing Wu, and Don Yee San Francisco Estuary Institute, Richmond, California On December 15, 2015 For Regional Monitoring Program for Water Quality in San Francisco Bay (RMP) Sources Pathways and Loadings Workgroup (SPLWG) Small Tributaries Loading Strategy (STLS) SAN FRANCISCO ESTUARY INSTITUTE | AQUATIC SCIENCE CENTER 4911 Central Avenue, Richmond, CA 94804 • p: 510-746-7334 (SFEI) • f: 510-746-7300 • www.sfei.org THIS REPORT SHOULD BE CITED AS: McKee, L.J. Gilbreath, N., Hunt, J.A., Wu, J., and Yee, D., 2015. Sources, Pathways and Loadings: Multi-Year Synthesis with a Focus on PCBs and Hg. A technical report prepared for the Regional Monitoring Program for Water Quality in San Francisco Bay (RMP), Sources, Pathways and Loadings Workgroup (SPLWG), Small Tributaries Loading Strategy (STLS). SFEI Contribution No. 773. San Francisco Estuary Institute, Richmond, CA. Final Report Executive Summary This report provides a synthesis of information gathered since 2000 on sources, pathways, and loadings of pollutants of concern in the San Francisco Bay Area with a focus on polychlorinated biphenyls (PCBs) and total mercury (Hg). Concentration and load estimates for other pollutants of concern (POC) are provided in the Appendix tables but not supported by any synthesis or discussion in the main body of the report. The PCB and Hg TMDLs for San Francisco Bay call for implementation of control measures to reduce stormwater PCB loads from 20 kg to 2 kg by 2030 and to reduce stormwater Hg loads from 160 kg to 80 kg by 2028 with an interim milestone of 120 kg by 2018. These are very challenging objectives given that the 2 kg PCB load allocation translates to a mean annual yield of 0.31 g/km2 for the free flowing areas downstream from reservoirs (6,650 km2), a mean annual concentration of 1.33 ng/L (assuming an annual average flow from small tributaries of 1.5 km3), and mean annual particle ratio of 1.4 ng/g of suspended sediment load (assuming an average annual suspended sediment load of 1.4 million metric t). Similarly for Hg, the 80 kg load allocation translates to a mean annual yield of 12 g/km2, a mean annual concentration of 53 ng/L, and mean annual particle ratio of 58 ng/g of suspended sediment load. Concentrations of these low magnitudes have been observed commonly in the Bay Area for Hg and yields of this low magnitude have been observed for PCBs and Hg. However, concentrations of PCBs at these magnitudes have only been observed in Marsh Creek (a more rural watershed) and particle ratios of PCBs and Hg as low as these have never been observed in the region. Given these and many other challenges, a small tributary loading strategy (STLS or “Strategy”) was written to help guide information development towards cost-effective implementation and address a more refined set of management questions (MQs): MQ1. Which Bay tributaries (including stormwater conveyances) contribute most to Bay impairment from pollutants of concern (POCs); MQ2. What are the annual loads or concentrations of POCs from tributaries to the Bay; MQ3. What are the decadal-scale loading or concentration trends of POCs from small tributaries to the Bay; and, MQ4. What are the projected impacts of management actions (including control measures) on tributaries and where should these management actions be implemented to have the greatest beneficial impact. The objective of the Strategy document was to present a planning framework for small tributary loads monitoring within the RMP that is consistent with and complemented by monitoring for POC and sediment loads that were completed in compliance with the Municipal Regional Permit (MRP) for MS4 Phase I stormwater agencies (SFRWQCB, 2009). The Strategy laid out a general series of tasks including dynamic and spreadsheet watershed modeling, planning studies to support monitoring design, information development for source area identification and monitoring, and small tributaries loads monitoring, where some were planned for earlier effort and others were placeholders to be picked up with more effort later. Successive updates of a multi-year plan (MYP) described progress and updated 3 of 100 Final Report priorities. This report synthesizes results from efforts carried out in relation to the Strategy and other relevant data sources to support a future update of the STLS in relation to a new MYP that will be issued in early 2016. During the first term of the MRP, permittees were asked to continue to improve information on locating high leverage tributaries (MQ1) using a combination of field monitoring and GIS information development. However, the bulk of the Regional Monitoring Program (RMP) STLS funding allocation was focused on better defining baseline concentrations and loads in six watersheds and regional loads by developing and using a GIS based regional “spreadsheet” model (RWSM). In addition, outside of RMP funding, the agencies that make up the Bay Area Stormwater Management Agencies Association (BASMAA) have been addressing MQ1 and MQ4 through sampling soils and sediments in old industrial areas to identify source areas with elevated concentrations, and testing and refining recommendations on which control measures may be most cost-effective. As a result of these efforts, stormwater characterization data now exist for PCB and Hg concentrations in over 25 watersheds, loads have been computed for 11 watersheds, and structural development of the RWSM has been completed. More refined estimates of regional loads using land use-based scaling of the climatically adjusted mean loads for each watershed support the use of 20 kg/y as a reasonable baseline PCB load (the starting point described in the TMDL before load reduction effort began), whereas 100-110 kg/y may be a better baseline load for Hg, although both new estimates may still be biased low if more polluted source properties, source areas, and high leverage subwatersheds are found in the future. A reanalysis of sediment and soil concentrations in industrial areas has helped to refine source areas, and work on identifying more high leverage tributaries is ongoing. Despite very good progress on MQ2 (improved estimates of regional scale and single watershed loads), only moderate progress has been made on MQ1 (Identification of new high leverage small tributaries) and MQ4 (Identification of areas and measures for cost effective management), and no progress has been made on MQ3 (Trends in concentrations and loads). There is presently no trend monitoring program in place to assess progress towards load reductions and improved environmental quality downstream, and knowledge about the performance of each management measure in relation to contamination levels in the landscape remains limited (MQ4). Thus areas for continued study have been identified, in the following order of priority. The first relates to supporting the identification of source properties, defined as those areas where focused application of clean up and abatement techniques could potentially cost-effectively remove large PCB masses (MQ4). The second relates to the ongoing need to find and characterize watersheds and subwatersheds with relatively high concentrations and particle ratios adjacent to sensitive Bay margin areas (high leverage areas) (MQ1) to indicate where further source identification work may be fruitful upstream (MQ4). In areas where moderate concentrations are found, the application of green infrastructure and other forms of redevelopment retrofit could possibly remove moderate amounts of PCBs cost- effectively with benefits for Hg and other POCs (although cost-effectiveness will continue to be low if these projects are implemented one by one). In addition to these two higher priority information gaps, the RWSM needs to be calibrated and published for general use with improved parameterization, accuracy and user flexibility. Once calibrated, the model can be used to predict baseline loads at any 4 of 100 Final Report scale (MQ2) and load reductions that might be achieved through management actions (MQ4). Lastly, the framework and design of a trends monitoring program needs to be completed (MQ3). Given the increasing focus on finding watersheds and land areas within watersheds at a scale paralleling management efforts and a transition from pilot-testing management measures in a few specific locations during the first MRP term to a greater amount of focused implementation and larger scale planning in the second MRP term, the following changes are recommended to the small tributaries monitoring and modeling program described by the STLS and supported by the RMP: 1. Cease fixed station loads monitoring. Instead, use a nimble watershed characterization monitoring design for identifying and characterizing concentrations in a greater number of watersheds and subwatersheds. The majority of the samples should be devoted to identifying areas of high leverage (indicated by high particle ratios or concentrations relative to other sites). In addition, a small number of monitoring sites should be allocated to sample potentially cleaner and variably-sized watersheds to help broaden the data set for regional model calibration and to inform decisions about cleanup potential. This sampling method directly addresses STLS MQ1 and also provides excellent data to support MQ2 and MQ4. 2. Develop a trends monitoring strategy that